Not so fast...
The experiment proceeds in three steps:
1) The magnet is held to prevent it from rotating, while the disc is spun on its axis. The result is that the galvanometer registers a direct current. The apparatus therefore acts as a generator, variously called the Faraday generator, the Faraday disc, or the homopolar (or unipolar) generator.
2) The disc is held stationary while the magnet is spun on its axis. The result is that the galvanometer registers no current.
3) The disc and magnet are spun together. The galvanometer registers a current, as it did in step 1.
Coils do not necessarily create a Lorentz Force just because there is Current cutting across a Magnetic Field. The Magnetic Field itself must have "movement" or the Current must vary.
This is the same problem I had with another idea. FEMM is a place to "start", but you have to be careful because it can give results that don't apply in real life situations.
The Faraday Paradox reveals WHY things developed the way they did.
Static vectors like with the Left Hand Rule:
...really only tell part of the story.
For the static vectors to apply there needs to be either variations in the magnetic flux or variations in the current flow or both. The Faraday Disk "works" only because as the copper disk rotates it's lines of current flow are changing. In a sense it's creating it's own "current wave" as the disk moves.
From this we can logically deduce the Induction motor because it's doing the same thing (creating a wave in a material) but it's just doing it better.
Permanent magnet motors are a natural logical leap too because the first thing you would dream up for the Faraday Disk is to use MULTIPLE current paths to create multiple waves in the disk.
...but once you've gone this far you think:
"Hey, if rapidly changing fields make things better why not alternate fields?"
This means that the Ring Magnet Motor "could" be made to work, but the coils cannot be continuous. It would be necessary to partition the coils so that they turn on and off. Timing is not too important compared to a brushless permanent magnet motor.
So the Ring Magnet Motor can be made to work... but for the effort required you are better off choosing the superior design which is the alternating magnetic field permanent magnet motor. However, the elimination of timing requirements means that if coil rotation is achieved (somehow) you don't need hall sensors. The coil rotation would probably best be tied to current levels so that higher currents translate to quicker rotation... but like sensorless RC ESC's they can operate independently.
The biggest hoped for benefit of the Ring Magnet Motor was it's simplicity, but the Faraday Paradox removes that from us and that probably knocks it out of contention.
So the Faraday Paradox actually eliminates a whole dimension of potential design concepts. One must construct any motor so that motion travels across flux lines.
One cannot design a motor so as to have motion that would go in and out relative to the cylinders in the image above. It could work if you used spheres because in the third dimension there would be flux lines that you would cross through.
Unless the cylinders were "infinite" in depth you will have a change as you went inward, so in practice most magnets will not violate the Faraday Paradox, but the Ring Magnet in effect is an "infinite" object. So some discretion is advised.
Generally it's best to have "hard flux line crossings" because that's how Forces become the strongest.
Combine crossings (in a powerful field) with current reversal and you have the best of both.
This does tend to point to the inevitability of alternating poles.
Back in 2006 when I first got "hooked" on ebikes I first was exposed to the CSIRO solar car that was built around Halbach Arrays. I knew then that as far as all the professionals were concerned this was the design they ended up with and that's generally a clue that it's well thought out.
The implementation of the Halbach Arrays on the CSIRO were competely incompatible with the needs of ebikes and the existing solution of 25 lb hub motors was just too ugly to accept, so I started thinking about alternatives.
For a couple of years on "endless sphere" I argued that multispeed gearing running through the conventional derailler was a very good idea because the legal 750 watt (or 250 watt) motors could be used over a wider set of conditions. I seem to have won that debate because mid drives seem to be blossoming everywhere you look. Who would have imagined?
<<< Seeking "radical alternatives" to the already improved mid drive gets harder. >>>
By "thinking outside the box" you have the potential to make a wrong turn and get "lost". The Faraday Paradox is an example of being lost in the woods.
Being humbled is good.
The Halbach Array is still the best possible way to generate power because it offers all the best features:
1) Magnetic field strength and polarity switches "violently" back and forth. (good)
2) Coils can exist without silicon steel which eliminates cogging and hysteresis losses.
3) Current switches directions as "violently" as the magnetic field. (good)
...the trick would be to implement those ideals on an ebike effectively and since the laws prevent high power levels you mostly want to stay with a mid drive.
The Halbach Solenoid as a concept?
Still not sure.
Was thinking about possible metaphors to highlight the Faraday Paradox and I thought about wood.
Wood has grains that run in the long direction and rings that run across it's width.
In order to generate a Force in a magnetic field using a current you in effect need to always "cut across the grains".
If you "cut along the grains" (ripping lumber) then no Force is created because you are not cutting across as many ring lines.
Some types of wood have very irregular grains and while you can be guaranteed that you will cut across SOME ring lines you also get a lot of chaos.
In a motor we want really straight lines and "sharp cuts across the grain".
However... if you can think in weird shapes that you are able to slice through it could be possible to come up with other designs. Straight lines and sharp cuts are easier for our minds to comprehend, but we only limit ourselves if we choose to.
The universe allows us to know just enough to be dangerous. ;)
I was thinking of "non-linear" means of crossing flux lines and thought about a circle rotating about another circle.
So I'm thinking if you placed a disc magnet in the center and had some iron in place to spread the flux around you would now have a homopolar motor "base" similiar to the Faraday Disk which brings about the Faraday Paradox.
Where it gets fun is when the coils oscillate (somehow with a crankarm system) so that they are "cutting across the grain" of flux lines. This gets around the Faraday Paradox.
If this could be figured out correctly then you have a motor without phase transitions that moves in and out of the flux path spinning in just one direction. No timing. No switching. Current in one direction spins the motor one way only. It looks like you should be able to get at least 25% copper utilization which it's that bad.
Hey... it's just a random thought at the moment. ;)
The old Wankel engines did pretty well.
I know that late in the career of Klaus Halbach he became interested in "magnetic lensing".
What he started doing is using odd shapes to bend magnetic flux into all sorts of strange ways.
The "Halbach Array" is what he will be remembered for, but that's very square stuff.
I stumbled upon something weird. In most cases when you have an air gap the natural tendency is for magnetic flux to jump straight across because it follows the shortest path.
In the image above the flux lines are not parallel with the normal direction.
This could lead to designs where flux is purposely bent.
Note: "NOT Parallel" is probably a bad choice of words because the lines actually are parallel with each other, but are running at nearly a 45 degree angle from what you normally experience. Whatever you call it I've never seen this sort thing at this scale and flux level before.
Just more about the Faraday Paradox.
If you looked at it in terms of orbits you can spin in one direction and never cross any of the flux lines, but if you went in most other orbits you would cross lines.
So you simply need to remember that not all paths are possible.
A low orbit like that passes through very high flux areas (in absolute value), but without crossing any flux lines you don't get any Force out of a current.
It means you can't think of flux as a separate independent entity, it's all about flux transitions.
I'm still thinking an eliptical orbit could work, but it's also possible that Forces can cancel each other out when you get into the gritty details. (haven't gotten that far yet)
In a "normal motor" the magnetic field is varied by having alternating polarity in a segmented architecture. (a permanent magnet motor for instance)
What I'm thinking is to oscillate (wobble) a non-alternating field that is either shaped like a Valley (image above) or like a Hill.
The coils are of both positive and negative, but they PHYSICALLY are aligned by mechanical movement so that (in the case of the Valley) they only push "downhill".
It's a homopolar motor but the field wobbles. (still fiddling around with a working design)
I find that by the time I truly understand something that I'm then able to know "how" to search for it on the internet. At this point the "idea" is so vague that I can't find out if it's right or wrong yet!
So the thinking is to have two circles "interact" in such a way that forward motion is possible without resorting to alternating current. So "homopolar" in the sense that there is nothing but forward current, but the design might actually involve alternating magnetic polarity and/or flux gradients.
Those "gray areas" would be a waste, so it would be nice to figure out how to eliminate them from the interactions between circles. (so a sparse pattern possibly)
...if you've ever been to an amusement park this ride gives some idea of the unexpected patterns that can occur using only circles.
So this is the kind of pattern that would match a motor:
What's interesting is that the magnetic flux flows in a circle, but because of mechanical motion the path allows the coils to cut across flux lines.
Circles normally do NOT have flux lines that are crossable.
...I'm thinking "Halbach Array Principles" could apply.
It's a little like building a motor around a planetary gear, which for ebikes is good because we want that geardown functionality built in. (we want it geared down)
This goes off on a tangent from your current train of thought, but what about using piezoelectric effect for generating motion? It would essentially act like a stepper motor so it could run slowly without loosing efficiency, couldn't it?
With a hammer, a chisle can be made. With a hammer and chistle, files can be made. With hammer, chisle, and file anything else can be made.
Kaishan k500w retired, Merida TEV500 on indefinite sabbatical, currently using a Currie E-zip Trailz.
How much power potential does that effect have?
I really don't know about it... (off the top of my head)
...just how much power can you get from a crystal?
And how big might that crystal have to be to produce 750 watts of power? (weight issues)
Could it run at 90% efficiency?
Is it something that is potentially affordable at our scale?
(I figured it was for very small devices)
Which Halbach Array is better, Squares or Triangles?
I say this as a joke obviously. The advantage of Squares is that the peak magnetic flux is higher and it's entirely linear, so if you are using a low inductance coil (very likely) then it makes sense to use Squares. However, if you wanted to increase the inductance and soften the phase transitions on the electronics (which really get worked hard when inductance is low) then the Triangles make for a smoother curved shape.
They both deliver the same performance as an absolute value... just different shapes.
There are few true Sine Wave controllers to pick from, so most people would be better served with Squares given the available controller options. So the default choice is Squares.
Another oddity I stumbled across is using Triangles to produce a zig zag pattern of flux.
Honestly I don't see much use for it, but it does show that a net unidirectional flow can be achieved. The problem is that based on how the Force vector relates to the Magnetic Flux vector there's no way to use it.
They say "a picture is worth a thousand words."
If that's true then an animation is worth a million.
Basically I'm trying to imagine some way to shape magnets around a planetary gear so you get the geardown built right into the motor and (hopefully) also make it work in a homopolar way. (without current switching)
The short answer is I don't actually have an answer... so that's really short.
But I do have the question... and if you pursue those you can actually get answers sometimes.
If you create a Halbach Cylinder using cylinder shaped magnets you could rotate them in a planetary gear kind of way and the coils (both positive and negative) would not need to move because the magnetic field would alternate around it. In the animation above the cylinder magnets rotate, but if you made them circle like a planetary gear then the arrow would stay in the same position. So it "appears" you can combine Halbach Cylinders, Planetary Gearing and Homopolar (one direction) current into a motor.
For the rotating cylinder magnets North is:
Blue - Up
Red - Left
Green - Right
Tan - Down
I quote from Wikipedia:
"Rotating the rods relative to each other results in many possibilities including a dynamically variable field and various dipolar configurations."
Yes indeed... this is true...
I'm just not seeing the Torque I need.
Basically I would need to spin the rotating magnets at 1000 rpm about the coils (positive and negative) before I'd be getting real power.
The central idea of this exercise is to design a 100 rpm motor without geardowns.
The solenoid type designs actually do better with linear Forces.
Basically for a given amount of weight in a compact space you achieve torque through rpm and gear reduction... which is where I started.
I do like the fact that a motor could be created that doesn't need polarity switching (homopolar) but having little magnets spinning around at 1000 rpm is excessive. That's introducing it's own friction and reliability problems.
Fun to learn new things anyway...
What I really need is more room to spread out. A maximum radius of 3" just isn't enough space at 100 rpm to generate the power you need. (unless it's 4" wide and weighs a lot)
If some kind of Halbach Array could travel along a conveyor belt system you might be able to hide it inside the downtube.
The bottom line is a Rotating Halbach Cylinder produces the same power as a conventional motor... you just do it in a slighly different way. (only one way current, no switching)
Power is equal to Torque times the Rotational velocity.
These small motors are like "little girls" lifting their light weights hundreds of times in a Jazzercise class. YES, they do get a workout, but the amount they lift is small.
They don't have enough muscle.
In order to SLOW things down we need to increase TORQUE.
These Ring Magnets can be bought for $16 each:
...so now I'm after just plain raw muscle. Finesse doesn't do it because ultimately the "Power equals Torque times Rotational velocity equation" forces you this way.
There really "isn't" a choice if speed reduction is the goal.
Note: The animation pulls 200 Newtons for 400 watts... that's real muscle.
If you aren't afraid of hurting yourself during assembly then this is even stronger:
...a 300 lb pulling Force for a physical weight of less than 2 lbs. And besides, fingers are overrated, who really needs them? (kidding)
One cannot do an Oktoberfest without an Accordion.
So I was thinking about how the Accordion works. The material expands and contracts like the piezoelectric crystals do, but on a large scale.
I have no idea how to do this correctly right now... but the piezoelectric "hint" may have helped.
Now if we can get the world to sing:
Ein Prosit, Ein Prosit, der Gemütlichkeit
Ein Prosit, Ein Prosit, der Gemütlichkeit
...then it will become a better place. :)
The obvious advantage is that long stroke piston action becomes possible because the Accordion-like behavior allows unlimited small air gaps rather than just one.
Just when you thought you were safe from "perpetual motion machines" we are introduced to "Primer Fields". (kidding)
Actually, when you create the bowl shaped magnets in FEMM you recreate what the guy is talking about, so it does seem to be founded on actual realities. We all are seeking the "perfect model" of the universe and as I previously experienced while using Vectors "alone" for motor models the Faraday Paradox proved that trying to see things that way is inadequate.
Search on google for "Primer Fields" and check out his youtubes.
If nothing else it's thought provoking.
Heck, we can't even reconcile quantum mechanics with relativity, so we really don't understand anything yet, so he's giving it his best shot with "Primer Fields".
The "real reality" will probably not be known in my lifetime.
Piezoelectric idea NOT a likely possibility using Crystals.
Conversely, those same crystals will change about 0.1% of their static dimension when an external electric field is applied to the material.
...so a one meter long crystal only changes length by one millimeter.
However, an Accordion style magnetic system "could work" so the idea was worth investigating.
A longer stroke allows for greater leverage on the crank and that means more Torque. We want the maximum Torque possible at the slowest speed. (100 rpm maximum)
In a typical solenoid there is just one rather large gap. This gap is at most a quarter of an inch or so because beyond that and magnetic field strength drops off.
Also, many solenoids use a return spring.
So I'm thinking you alternate "spring washers" (to get the rebound you need) and silicon steel flat washers to carry most of the magnetic flux.
Whatever is required to get the stroke you want.
Meanwhile the whole thing is wrapped with a coil and outside that a silicon steel tube which doubles as the bikes downtube.
Will need to do a FEMM simulation.
If not springs then opposing magnets might work if they are just strong enough to create a rebound. Too much magnet would be bad.
The idea of a fine grained "Accordion" type device with a long stroke is very attractive. A long stroke creates better crank leverage and more Torque without forcing you to increase rpm.
The length of the magnetic flux return path is a bit extreme however. Not sure if it might not be better to break things up into bite sized chunks.
I've sort of neglected the most obvious idea.
You could always use a short stroke solenoid (or two) with a linkage that magnifies the stroke through leverage.
An inch of stroke could easily become three inches.
The advantage is that you can focus the generation of Force through magnetic methods into the most concentrated space. A strong but lightweight linkage (aluminum) could then connect to the crank which then connects to the bottom bracket. You are still looking at just a few bearings overall, much less than multiple geardowns.
Many bicycles use shock linkages... so that's an example.
Obviously you don't "create" additional power by using leverage, but you do make things more manageable as far as stresses on your components. By gaining a 3-to-1 ratio with a lever you lose a 3-to-1 ratio as far as Force applied, but you do reduce the "slop" in the bearings and reduce the potential of breakage.
Appropriate for Valentines Day I think?
Actually what is interesting is that the magnetic flux increases linearly as you move towards the center of the heart.
The shape encourages a longer stroke and deeper penetration... which is also probably a good thing for Valentines Day. ;)
But then I realize that if you switch positions you can alter the magnetic flux from something very linear and rising, to something nearly flat and constant.
So it appears that there is considerable flexibility in the design to be able to "shape" the power you can get from a longer stroke solenoid.
You can get very creative with this.
Winter is easing and Spring is near.
Contemplating deep thoughts in Winter is great entertainment, but with Spring coming it's probably time to put this thread into "park".
Go outdoors... ride a bike... have fun... and be friendly.
(this probably feels early for those on the East coast right now)
I'm still getting Spring Fever, but I noticed there was some activity on Endless Sphere along similiar lines.
What if you did one of these wheels without spokes and hubs, but you went ahead and inserted a Brushless Permanent Magnet Motor into that wheel?
The primary criticism of these "spokeless wheels" is that an increased amount of weight is required in the rims and that is bad for handling and inertia. But if you are already throwing a lot of weight into a rim (because you are making it into a motor) then a normally bad design can become a pretty good one.
I kind of think at the end of the day the mid drives are essentially unbeatable. Maybe a solenoid type crank drive could make sense too, still uncertain on that. Hub motors are just too heavy. A Switched Reluctance Rim Motor has potential if you build it into a carbon fiber rim using a minimum amount of steel filings embedded into the resin. Generally the small motors that spin fast get more "bang for the buck" but at the cost of reliability. Professionals that produce high precision mid drives will probably be victorious in the end. But we still love trying alternatives.
Okay... back to vacation...
Read it here for yourself folks what granpa safe alludes to & is the endlessss source for ALL of his ideas.
We've all been repeating the same basic ideas over and over again since about 2003 when the first people started getting into this stuff.
I'd actually characterize the ebike as becoming more "mature".
Professional people are getting in on the idea and we are seeing really top notch mid drives now that very likely represent the "endpoint" of this long journey.
Am I suggesting I'm the originator of all this?
But to be historically accurate I did advocate multispeed gearing integrated into the powertrain (mid drive) since at least 2007. My first ebike used a standard bicycle derailler. (10,000 miles usage)
I'm seeing a sort of technological "endpoint" now where unless some idea comes forth that really changes the paradigm we will see a future largely populated with mid drives.
But you never know... it's just fun to imagine the ideas. :)
On a personal note:
I'm in my 50's now and way too old for the crazy risk taking I used to do. So I realized a few years ago that my health was more important that any thrills I can get from dangerous activities. I ride about two hours a day on a regular bike and as long as I'm healthy enough to keep doing it I'll just use human propulsion rather than electric.
At some point I might actually get old enough to actually need an ebike as a "helper", but that could be a decade or two away.
The EBRR ("Electric Bicycle Road Racing") on Go Kart tracks still seems like a great idea and if people all adopted some standardized mid drive it could become a nice new sport. The last crack at the idea failed because the mid drive hadn't really arrived yet and people confused the issue by bringing ebikes with motorcycle level power. (making the whole experience absurd) Some younger generation can take up Go Kart track racing and just buy what they need now as mid drives flood the market.
1000 Watt Input Limit, Mid Drives, Go Kart Tracks
I actually did create this, but with only a single gear and "current limiting". It sits in storage.
Back to Spring Fever...
Speaking of accusing someone of stealing ideas:
I had done some similiar things years ago:
I actually like the idea, I just realized it was beyond my technical abilities to build and probably slightly ahead of my intellectual abilities at the time to configure properly.
The idea of blending such a motor in with the cassette is a great idea because it allows you to use existing bicycle parts. A motor that is "backwardly compatible" is a great idea.
Upon deeper analysis I realized that the "disc insertion" coils idea would not produce as much power as I wanted, so that was another reason I stopped working on it. It's a really, really difficult job to manufacture too.
In order to actually get "something running" on the bike project I ended up just dumping the disc idea and going back to the standard chain and sprockets. Keep It Simple Stupid is sometimes what you do when you need to finish a project.
Okay, hopefully I can now slip back into my Spring Fever Vacation... :)
There are currently 0 users online.
Support V is for Voltage
Disclosure: Monetized by Skimlinks
Communal learning about moving our butts around town without burning fossil fuels.